W3C home > Mailing lists > Public > www-html@w3.org > February 2001

RE: client side includes

From: Russell O'Connor <roconnor@math.berkeley.edu>
Date: Sun, 4 Feb 2001 13:23:30 -0800 (PST)
To: "'www-html@w3.org'" <www-html@w3.org>
Message-ID: <Pine.SOL.4.32.0102041316230.1395-100000@pub-708c-17>
On Sun, 4 Feb 2001, Murray Macdonald wrote:

> While it is true HTTP1.1 saves some time by not needing to continually
> reestablish connections, it remains true that UAs by default only have
> 5 or so connections open at a time.  Unless I am mistaken, even with
> HTTP1.1 a quantity of CSI files (images, objects, js includes, html
> includes) requires the time of multiple round trips.

Even if 30 CSI were realistic, I believe section 8.1.2.2 of HTTP 1.1
<ftp://ftp.isi.edu/in-notes/rfc2616.txt> on pipelining should address your
issue.  But I'm not all that well read on HTTP/1.1.

As a bonus, using entities for client side includes places the names of
the required entities to retrieve at the beginning of the document.  So a
UA can begin downloading the pieces to include even before the spot where
it needs to be included is retrieved.

-- 
Russell O'Connor
           <http://www.math.berkeley.edu/~roconnor/>
``Paradoxically, a refusal to `put a monetary value on life' means that
life is often undervalued.'' -- Artificial Intelligence: A Modern Approach
Received on Sunday, 4 February 2001 16:23:32 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:45 GMT